Nguyen The An

We are delighted that our group has 8 papers accepted at ICLR 2025. 🙌🙌 These are great work of the amazing students Tran Viet Hoang, Rachel Si Yan Teo (@rachtsy_ ), Laziz Abdullaev (@lazizabdulla ), Viet Dung Nguyen (@DungNv1714 ), Minh-Khoi Nguyen-Nhat (@_minhkhoi ), Stefan Nielsen (@svkn_1 ), An The Nguyen (@sg__nta ), Thanh Chu (@thchulz ), Tung Duy Pham (@pdt931604 ), Trang Pham (@huyen_tran67454 ), Tho Tran (@Tho_Tran99 ), Thanh Tran (@Thanhqt2002_GAR ), Minh Hoang Nguyen (@minhnh2127 ), Luc Nguyen, our exceptional research fellow, Dr. Thieu Ngoc Vo (@thieunvo ), and our incredible collaborators, Professor Xin Tong, Professor Tam Le (@TamTLe ), and Dr. Linh Duy Tran (@tran_linh87). Our accepted papers were made possible through the tremendous support from our Department of Mathematics at NUS and valued industrial collaborators, FSOFT AI Center and VinAI Research. We would also like to thank the reviewers and chairs at ICLR 2025 for their thoughtful consideration, valuable comments, and constructive feedback, which have helped us enhance our papers. 👏👏 🌟🌟 My student at NUS Math, Tran Viet Hoang, is the first author of three accepted papers, while Rachel Si Yan Teo and Dr. Thieu Vo Ngoc have two first-author papers this time. Additional big kudos to Viet Dung Nguyen, Minh-Khoi Nguyen-Nhat, An The Nguyen, Thanh Chu for their first first-author papers at the top ML conferences. I also have two more two-author papers with my students at NUS Math, Laziz Abdullaev and Rachel Si Yan Teo. 🚀🚀 Our papers focus on three key topics: Deep Learning Models (Sparse Mixture of Experts, Transformers, and Mamba), Tree-Sliced Wasserstein Distance, and Equivariant Neural Functional Networks. Below is a list of our accepted papers. 1. "Equivariant Neural Functional Networks for Transformers" openreview.net/pdf/859fa73e86… Hoang V. Tran, Thieu Vo, An Nguyen The, Tho Tran Huu, Minh-Khoi Nguyen-Nhat, Thanh Tran, Duy-Tung Pham, Tan Minh Nguyen 2. "Distance-Based Tree-Sliced Wasserstein Distance" openreview.net/pdf/34b36a5a43… Hoang V. Tran, Minh-Khoi Nguyen-Nhat, Huyen Trang Pham, Thanh Chu, Tam Le, Tan Minh Nguyen 3. "Spherical Tree-Sliced Wasserstein Distance" openreview.net/pdf/166f61a052… Hoang V. Tran, Thanh Chu, Minh-Khoi Nguyen-Nhat, Huyen Trang Pham, Tam Le, Tan Minh Nguyen 4. "Demystifying the Token Dynamics of Deep Selective State Space Models" openreview.net/pdf/5b044a47c5… Thieu Vo, Duy-Tung Pham, Xin T. Tong, Tan Minh Nguyen 5. "Transformer Meets Twicing: Harnessing Unattended Residual Information" openreview.net/pdf/949fd5669f… Laziz Abdullaev, Tan Minh Nguyen 6. "MoLEx: Mixture of Layer Experts for Fine-tuning with Sparse Upcycling" openreview.net/pdf/6f74492690… Rachel Teo, Tan Minh Nguyen 7. "CAMEx: Curvature-aware Merging of Experts" openreview.net/pdf/89e794f538… Viet Dung Nguyen, Minh Nguyen Hoang, Rachel Teo, Luc Nguyen, Tan Minh Nguyen, Linh Duy Tran 8. "Tight Clusters Make Specialized Experts" openreview.net/pdf/56cac7674a… Stefan Nielsen, Rachel Teo, Laziz Abdullaev, Tan Minh Nguyen See you all in 🇸🇬! @iclr_conf, #iclr, #ICLR2025, #MixtureOfExperts, #transformer, #SelfAttention, #optimaltransport, #equivariant, #LLM, #TrustworthyAI, #RobustAI, #AIsafety, #MLTheory, #DLTheory, #AI, #MachineLearning, #DeepLearning